Skip to content

Data sources

Per-source ingestion specs — the Josh-specific detail for each source: the schema we land, chunker choice, citation ID format, status, and probe findings. Each page links out to the upstream API documentation for endpoint shapes and field definitions rather than reproducing them.

For the canonical inventory and v1/v2 status, see the source inventory. For what's in the substrate right now — row counts, last-run freshness, ingestion errors — see data status.